mzbac/linkedin-message-assistant

Linkedin-AI-Writer

A Chrome extension integrated with your open source LLM to help you write better Linkedin messages.

Installation

• Download chrome.zip from Releases.
• Unzip the file
• In Chrome/Edge go to the extensions page (chrome://extensions or edge://extensions).
• Enable Developer Mode.
• Drag the unzipped folder anywhere on the page to import it (do not delete the folder afterwards).
• Click on the extension icon.
• Paste your local LLM url (e.g. TGI endpoint http://localhost:8000/generate or Llamacpp endpoint http://localhost:8000/completion) and click on Save.
• Open Linkedin message page in your browser.
• Type your objective or intention of the message in the text box and click the button at the bottom to generate the message.
• Enjoy!

Development

Install npm packages

npm install

Build the extension

npm run build

Limitations

• Currently, only the dolphin-2.2-yi-34b model prompt format is supported due to it being the best model I found for writing LinkedIn messages assistant.

• The backend API only supports HuggingFace text generation inference and llama.cpp server. You can lauch the backend as shown below:

  docker run --gpus all --shm-size 1g -p 8000:80 -v $PWD/models:/data ghcr.io/huggingface/text-generation-inference:latest --max-total-tokens 2050 --max-input-length 1024 --max-batch-prefill-tokens 2048 --quantize awq --model-id TheBloke/dolphin-2_2-yi-34b-AWQ

  #!/bin/bash
  
  # Define the repository and model details
  REPO_URL="git@github.com:ggerganov/llama.cpp.git"
  REPO_DIR="llama.cpp"
  MODEL_URL="https://huggingface.co/TheBloke/dolphin-2_2-yi-34b-GGUF/resolve/main/dolphin-2_2-yi-34b.Q4_K_M.gguf"
  MODEL_FILE="dolphin-2_2-yi-34b.Q4_K_M.gguf"
  
  # Clone the repository if it doesn't already exist
  if [ ! -d "$REPO_DIR" ]; then
  git clone "$REPO_URL"
  fi
  
  # Change directory to the cloned repository
  cd "$REPO_DIR"
  
  # Build the project using make
  # Assume make is idempotent, as it should only rebuild changed files
  make
  
  # Change directory to the models folder
  mkdir -p models
  cd models
  
  # Download the model if it doesn't already exist
  if [ ! -f "$MODEL_FILE" ]; then
  wget "$MODEL_URL"
  fi
  
  # Change directory back to the project root
  cd ../
  
  # Launch the server
  ./server -m models/"$MODEL_FILE" --host 0.0.0.0 --port 8080 -c 16000